5.1. Role of Platelets

In addition to their role in primary hemostasis, activated platelets provide an efficient catalytic surface for the assembly of the enzyme complexes of the blood coagulation system, also known as secondary hemostasis. The classic description of coagulation involved a cascade model consisting of two distinct pathways: the extrinsic, or tissue factor pathway and the intrinsic pathway. These are now viewed in terms of overlapping phases of initiation, amplification and propagation (Figure 5.1). These pathways have been reviewed in detail elsewhere [172–174]; we will describe them briefly and focus on the interactions between platelets and blood coagulation.

Figure 5.1

Simplified model of the major enzymatic processes of the coagulation cascade. In this model, coagulation is initiated by binding of tissue factor to factor VIIa; this complex activates Factor IX and Factor X, and converts more Factor VII to VIIa. Factor (more...)

Tissue factor binds factor VIIa, which exists in low levels in the blood, and this interaction converts more factor VII to VIIa autocatalytically. More importantly, the tissue factor-VIIa complex formed activates both factor IX and factor X. Factor IXa, in the presence of factor VIIIa, catalyzes the generation of more Xa, hence amplifying this pathway. Furthermore, factor Xa can also activate factor VII to VIIa. Tissue factor pathway inhibitor (TFPI), a multivalent, plasma protease inhibitor inhibits activated factor Xa, in a factor VIIa-dependent manner providing a feedback inhibition of the tissue factor pathway. Factor Xa, in the presence of Va and Ca²⁺ catalyzes the conversion of prothrombin to thrombin on anionic phospholipid surface. Thrombin converts fibrinogen to fibrin in a multiple-step reaction, which eventually leads to formation of cross-linked and insoluble interconnecting networks of strands. In addition, thrombin activates coagulation factors V, VIII and IX to generate active forms of Va, VIIIa and IXa. Thrombin is also key mediator of platelet activation, release reaction and aggregation. Its action on platelets produces a highly efficient catalytic surface for further generation of thrombin. The coagulation system consists of a number of serine proteases, cofactors, calcium, and cell membrane components, and their reactions are frequently illustrated as simple sequential events. However, they represent highly complex interactions, subject to regulation at a number of levels.

The characteristics of the procoagulant activity on the platelet surface resemble those of synthetic phospholipid surfaces. To be active in coagulation, the phospholipid surface requires a net negative charge (provided by phosphatidylserine, phosphatidylinositol or phosphatidic acid), an optimal degree of unsaturation of the acyl chains, and an appropriate size. The most optimal procoagulant activity is observed when the phospholipid is in the micellar form with an anionic phospholipid concentration of about 20–30%. Phospholipid surfaces accelerate two distinct reactions in the coagulation cascade, activation of factor X by factor IXa (in the presence of factor VIIIa and Ca²⁺) and activation of prothrombin by factor Xa (in the presence of factor Va and Ca²⁺), by providing sites for the assembly of enzyme substrate complexes [175].

In resting platelets, phosphatidylserine and other anionic phospholipids are located on the inner aspect of the membrane bilayer [176]. Following platelet activation with thrombin, collagen, or shear stress, phosphatidylserine moves from the inner to the outer leaflet of platelet plasma membrane. This movement is associated with an increase in the activation of prothrombin and factor X [177] and with the appearance of high-affinity binding sites for factors Va and VIIIa.

The importance of the exposure of anionic phospholipid for hemostasis in vivo is exemplified by Scott syndrome, a rare bleeding disorder first described in 1979 by Weiss et al. as an isolated deficiency of platelet procoagulant activity [178]. The patient had a bleeding disorder, including excessive postoperative bleeding and spontaneous retroperitoneal hematomas. Her platelets aggregated normally to all agonists tested but failed to provide procoagulant activity. In a series of subsequent studies, it was shown that her platelets had a reduced number of binding sites for factor Va and VIIIa and did not promote prothrombin or factor X activation. Furthermore, following platelet activation with thrombin and collagen, there was a marked decrease in the exposure of anionic phospholipid at the platelet surface as compared with normal platelet [179]. The clinical features of this patient illustrate the importance of platelet procoagulant activity in secondary hemostasis.

The transbilayer movement of anionic phospholipid from the inner leaflet to the outer leaflet requires increases in intraplatelet Ca²⁺⁺ and a putative enzyme named scramblase. Externalization of anionic phospholipid in platelets is accompanied by the generation of phosphatidylserine-rich microvesicles, which as far back as 1967, were shown to account for the clot-promoting activity of plasma [180]. By freeze-fracture electron microscopy they were shown to possess bilayer structure with intramembranous particles [181]. These microvesicles also contain platelet cytoskeletal proteins and platelet membrane glycoproteins Ib, IIb, IIIa and IV. The discovery of Sims et al. that microvesiculation is also defective in Scott Syndrome suggested that microvesicles and anionic phospholipid exposure are linked events and that failure to produce microvesicles may contribute to the hemostatic defect in Scott Syndrome [182].

Several findings suggest that, in addition to its role in normal hemostasis, platelet microvesiculation may contribute to the prothrombotic tendencies observed in several diseases. Platelet-derived microvesicles have been detected in the circulation in patients with disseminated intravascular coagulation [183], heparin-induced thrombocytopenia [184], the antiphospholipid antibody syndrome [185], transient ischemic attacks [186], and thrombotic thrombocytopenic purpura [187], conditions associated with either arterial, venous, and/or microvascular thrombosis. These associations suggest that, while they may be necessary for normal hemostasis, elevated microvesicle concentrations could predispose to thrombosis. Thus, conditions that increase the production or decrease the clearance of microvesicles are expected to increase the incidence of thrombosis. More recently, in addition to their hemostatic role, platelet-derived microvesicles were shown to stimulate hematopoietic cells [188] and to transfer platelet-specific receptors to the surface of other cells [189]. Microvesicles localize to sites of blood vessel damage [190] and they are incorporated to a growing thrombus [191]. In a mouse model of hemophilia, generation of microvesicles through an interaction with soluble P-selectin and PSGL-1 significantly improved the kinetics of fibrin formation in mice and normalized the prolonged tail-bleeding time [191].

Recent findings indicate that in addition to providing anionic phospholipid for prothrombin and factor X activation, microvesicles may have a major role in the initiation of thrombosis via the tissue factor pathway. These data also reveal that microvesicles relevant for thrombosis are derived from various cell types. The traditional concept of hemostasis has been that blood clotting is induced by tissue factor, a cell surface glycoprotein present on most tissues, which is derived from extravascular tissue following vessel wall injury [192]. Several studies have shown that tissue factor also circulates in the blood under normal conditions associated with cell-derived membrane microvesicles [193]. Tissue factor-bearing microvesicles are believed to arise from cells of the monocyte/macrophage lineage [194]; the role of leukocyte-derived tissue factor is discussed in greater detail in Section 5.2. The microvesicles bind activated platelets in thrombi through a molecular bridge between P-selectin glycoprotein ligand-1 (PSGL-1) on microvesicles and P-selectin on platelets. These microvesicles selectively enriched in both tissue factor and PSGL-1 fuse with activated platelets, transferring tissue factor to the platelet membrane [194]. Failure of this hemostatic mechanism may explain the ability of agents that block the PSGL-1–P-selectin interaction to significantly inhibit experimental thrombosis [195]. The role of tissue factor-bearing microvesicles in normal hemostasis is an active area of investigation.

5.2. Role of Platelet-Leukocyte Interactions

Leukocytes are key mediators in the inflammatory response, and interactions between leukocytes and platelets are increasingly recognized as relevant for inflammation and thrombosis [196]. Figure 5.2 depicts several proposed mechanisms by which platelet-leukocyte interactions contribute to thrombosis.

Figure 5.2

Schematic of interactions between leukocytes (neutrophil on left, monocyte on right), and platelets in activation of coagulation. The figure depicts three of the various reported mechanisms involved in platelet-leukocyte adhesion (mediated by platelet (more...)

The role of leukocytes in promoting platelet adhesion to endothelium was reviewed in Section 3.3; in addition, leukocyte-platelet interactions contribute to activation of the coagulation cascade via tissue factor-dependent pathways. While tissue factor antigen has been measured in normal plasma, it was assumed to be present in an inactive, or encrypted, form. However, a report by Giesen et al. [193] challenged this notion and prompted a re-evaluation of the role of blood-borne tissue factor in thrombosis. Those authors demonstrated that perfusion of human blood over pig arterial media (which had no detectable tissue factor) and over glass slides coated with collagen, resulted in formation of thrombi that stained intensely for tissue factor. Because the perfusion of blood was very brief (5 minutes), the authors proposed that blood-borne tissue factor mediated propagation of thrombi at sites of vascular injury. In these experiments, tissue factor was present on vesicles, which stained positively for CD18, a leukocyte marker, suggesting a leukocyte origin for these vesicles. Based on these findings, leukocyte-derived, tissue factor-bearing microvesicles are suggested to play a role in hemostasis and thrombosis. Recent data reveal that microvesicles derived from monocytes, originating from cholesterol-enriched membrane microdomains known as lipid rafts, can fuse with the membrane of activated platelets, via a mechanism involving P-selectin–PSGL-1-interactions [197]. These findings were proposed as a mechanism to explain limiting of coagulation to the sites of vascular injury, where activated platelets accumulate.

While resting monocytes contain minimal tissue factor, its synthesis is enhanced by a number of inflammatory mediators, including endotoxin, interleukin-1β, antigen–antibody complexes, tumor necrosis factor-α (TNF-α), platelet membranes, C-reactive protein, reactive oxygen species, and complement factor 5a [198]. In addition to synthesizing tissue factor, activated monocytes can activate, or decrypt, tissue factor. Like many other cells, stimulated monocytes can release microvesicles [199], though the precise mechanism of this process is not fully understood. Microvesiculation is associated with transbilayer movement of anionic phospholipids, from the inner to outer membrane bilayer, thus enabling decryption of tissue factor by these microvesicles. In addition to monocytes, neutrophils may also express tissue factor, either by de novo synthesis or via uptake of tissue factor-containing vesicles from plasma.

Leukocytes may contribute to thrombosis via other mechanisms besides tissue factor; for example, the leukocyte integrin Mac-1 can bind coagulation factor X, and mediate its activation by release of cathepsin G, a serine protease contained in leukocyte granules [200]. Further, neutrophil release of elastase and cathepsin G can cleave the platelet integrin GPIIb/IIIa at the carboxyl terminus of the α_IIb chain, resulting in a conformational change, increasing the capacity of platelets to bind fibrinogen [201]. Additional proposed mechanisms linking leukocytes and platelets to thrombosis have been reviewed in detail elsewhere [196,202].

A number of clinical observations demonstrate a significant association between elevated blood leukocyte count (leukocytosis) and worse clinical outcome. For example, in patients with myocardial infarction, leukocytosis was found to correlate significantly with recurrence of a coronary event, despite adjusting for other well-known risk factors [203]. Similarly, in a recent review of published studies involving >350,000 patients, a significant relationship between leukocyte count and morbidity and mortality was identified [204]. Whether these findings indicate a causal role for leukocytes in the thrombotic complications, or primarily reflect a greater inflammatory response in patients with more severe disease remains to be determined conclusively.